We have a usecase wherein a single "avatar" may have many hundreds (or even thousands) of skins. Think hats, shoes, etc.
The initial iteration will have only several of these. Later iterations will add more and more. More than 1 designer will be working with this "avatar."
Right now, it seems to us that the actual spine file can only be saved as a binary. This means that our source control can only function on it as a "single source of truth" that must be worked on by only one designer at the time. That's the first blocker for us - if there was some way to have multiple designers working on it at once and "merge" their changes, it'd be great.
The second thing we're trying to figure out is, if indeed the Spine file can only be used as a binary, how can we ensure that one designer didn't accidentally delete a slot, skin, texture, whatever, in their work on a new skin/texture/whatever?
We're thinking we could have as part of the commit process an automated build script on the server that runs a pre-configured build on the spine file to output the atlas, json, and png spritesheet, and then run tests to make sure everything we expect to be in the json is there. However, we aren't aware of any sort of programmatic API for the Spine editor that would allow us to hook in. For example, if we could do something like, from bash, spine build -c config.json documents/spine_avatar.spine > python test_spine_output.py
that would be dope.
What methods have other people used for a designer test workflow for the Spine editor?